<ul><li>This paper investigates the challenge of establishing stochastic-like guarantees for learning from a stream of data that includes both unknown clean-label adversarial samples and noise.</li><li>The approach allows the learner to abstain from making predictions when uncertain, and measures regret in terms of misclassification and abstention error.</li><li>The study corrects inaccuracies in the work of Goel, Hanneke, Moran, and Shetty and explores methods for the agnostic setting with random labels.</li><li>The paper introduces the concept of a clean-label adversary in the agnostic context and provides a theoretical analysis of a disagreement-based learner subject to a clean-label adversary with noise.</li></ul>

Adversarial Resilience against Clean-Label Attacks in Realizable and Noisy Settings

Discover more